[论文解读] Inherent Tradeoffs in Learning Fair Representations
本文在统计公平性(statistical parity)约束下,建立了联合误差的理论信息论下界,证明当不同群体的基率不同时,任何公平分类器必然在至少一个群体上产生最小误差。本文引入了TV-Barycenter问题,通过线性规划高效计算该下界,并提出一种基于Oracle贝叶斯分类器的最优随机化分类器,表明公平性约束以可量化的方式内在地限制了准确性。
Real-world applications of machine learning tools in high-stakes domains are often regulated to be fair, in the sense that the predicted target should satisfy some quantitative notion of parity with respect to a protected attribute. However, the exact tradeoff between fairness and accuracy is not entirely clear, even for the basic paradigm of classification problems. In this paper, we characterize an inherent tradeoff between statistical parity and accuracy in the classification setting by providing a lower bound on the sum of group-wise errors of any fair classifiers. Our impossibility theorem could be interpreted as a certain uncertainty principle in fairness: if the base rates differ among groups, then any fair classifier satisfying statistical parity has to incur a large error on at least one of the groups. We further extend this result to give a lower bound on the joint error of any (approximately) fair classifiers, from the perspective of learning fair representations. To show that our lower bound is tight, assuming oracle access to Bayes (potentially unfair) classifiers, we also construct an algorithm that returns a randomized classifier that is both optimal (in terms of accuracy) and fair. Interestingly, when the protected attribute can take more than two values, an extension of this lower bound does not admit an analytic solution. Nevertheless, in this case, we show that the lower bound can be efficiently computed by solving a linear program, which we term as the TV-Barycenter problem, a barycenter problem under the TV-distance. On the upside, we prove that if the group-wise Bayes optimal classifiers are close, then learning fair representations leads to an alternative notion of fairness, known as the accuracy parity, which states that the error rates are close between groups. Finally, we also conduct experiments on real-world datasets to confirm our theoretical findings.
研究动机与目标
- 理解在现实世界数据分布下,分类中公平性(统计公平性)与准确性的基本权衡。
- 在施加公平性约束时,特别是当基率不同时,刻画跨群体的最小可能联合误差。
- 开发一种在获得贝叶斯分类器的前提下,实现最优准确性的算法,同时满足公平性约束。
- 探索在特定条件下,学习公平表示是否能导致准确性公平性(accuracy parity)等其他公平性概念。
- 为当受保护属性具有多于两个取值时,提供一种计算公平性-准确性权衡下界的有效方法。
提出的方法
- 利用Kullback-Leibler散度和群体基率,推导在统计公平性下,公平分类器的群体误差之和的信息论下界。
- 提出一种随机化分类器构造方法,该方法在获得群体特定的贝叶斯分类器的Oracle访问权限下,可同时实现最优准确性和公平性。
- 当群体多于两个时,将下界计算公式化为一个线性规划问题,即在总变差距离下的TV-Barycenter问题。
- 通过证明所构造的随机化分类器在给定假设下可达到该下界,从而证明该下界是紧的。
- 分析在何种条件下,学习公平表示可导致准确性公平性,即群体间误差率相近。
- 在真实世界数据集上对理论发现进行实证验证,确认了权衡关系并验证了下界的紧致性。
实验结果
研究问题
- RQ1当不同群体的基率不同时,任何满足统计公平性的分类器在跨群体中必须承受的最小联合误差是多少?
- RQ2我们能否构造一个在统计公平性下既最优准确又公平的分类器?如果可以,其成立的假设条件是什么?
- RQ3当受保护属性具有多于两个取值时,如何高效计算公平性-准确性权衡的下界?
- RQ4在何种条件下,学习公平表示会导致准确性公平性,而非仅统计公平性?
- RQ5近似公平性约束对可实现准确性的影晌是什么?这种影响如何量化?
主要发现
- 当不同群体的基率不同时,任何满足统计公平性的分类器都必须承受最小联合误差,从而确立了公平性与准确性之间的内在权衡。
- 群体误差之和的下界在信息论上是紧的,且可通过由Oracle贝叶斯分类器构造的随机化分类器实现。
- 当群体多于两个时,可通过求解线性规划问题(作者称之为TV-Barycenter问题)高效计算该下界。
- 当基率不同时,该下界严格为正,意味着公平性无法在不损失整体准确性的前提下实现。
- 如果各群体的贝叶斯最优分类器彼此接近,则学习公平表示可导致准确性公平性,即各群体的误差率相近。
- 在真实世界数据集上的实证结果确认了理论权衡关系,并验证了所推导下界的紧致性。
更好的研究,从现在开始
从论文设计到论文写作,大幅缩短您的研究时间。
无需绑定信用卡
本解读由 AI 生成,并经人工编辑审核。